Delivery-only kitchens and shared commissaries get pitched as the lightweight option. No dining room, no front of house, no washrooms for a hundred covers, cheaper rent in a worse location. All true, and all of it makes them sound cheaper than they usually are.
What actually happens is that cost moves. The savings are real on the front of house. The back of house gets harder.
What genuinely gets cheaper
Front of house disappears. No dining room finishes, no bar, no customer washrooms at restaurant fixture counts, no furniture, and far less of the design budget spent on how the room feels.
Location cost drops. Delivery-only does not need frontage, foot traffic or visibility, which opens up industrial and secondary units at meaningfully lower rent.
Occupant load falls. Fewer people in the building generally means simpler exiting and life safety than a space designed to seat a crowd.
Those are real savings and they are the reason the model exists.
What gets harder
Production density. A restaurant kitchen serves the room it is attached to. A delivery kitchen serves whatever volume the platforms send, and a commissary may be producing for several businesses at once. More equipment in less space means more exhaust, more make-up air, more electrical and more refrigeration per square foot than a restaurant of the same size.
Ventilation does not care that there is no dining room. It cares what is under the hood, and delivery kitchens tend to have a lot under the hood.
Storage. This is the one that surprises people most. A restaurant turns stock through a predictable service pattern. A production kitchen holds ingredients, packaging, and finished product simultaneously — and packaging for delivery takes far more volume than most operators plan for. Cold storage and dry storage both need to be larger than instinct suggests.
Order handoff. Couriers arrive constantly and need somewhere to be that is not the middle of the kitchen. That means a pickup area, a route in and out, and ideally a door that is not the same one deliveries use. Get it wrong and you have strangers walking through a production area all day, which is both a food safety issue and an operational one.
Waste. Higher packaging volume and higher production volume mean more waste, more often, in a unit that may have been leased on the assumption of light industrial use.
The multi-tenant question
A shared commissary — several businesses producing in one facility — is a different regulatory animal from a single kitchen, and it is the part most likely to be underestimated.
Each operating business generally needs its own standing with the health unit, and the facility needs to work in a way that lets several of them operate without cross-contamination. Practically that means:
- Separated or clearly allocated storage, cold and dry, per operator
- A cleaning and sanitation schedule that works across shifts and businesses
- Enough handwash and warewash capacity for concurrent use, not sequential
- Clear allergen handling between operators
- Scheduling, because two businesses wanting the same hood at the same time is a business problem, not a design one
This is worth discussing with your local health unit early. The requirements are specific to how the facility will actually be used, and it is far cheaper to design around them than to retrofit after a plan review conversation.
Zoning is the first question, not the last
A delivery kitchen in an industrial unit is a food premises in a zone that may not contemplate one, and a commissary producing for other businesses may read as food manufacturing rather than a restaurant. Those can be different permitted uses.
Confirm the specific use is permitted at the specific address before signing. This is true for every food project, but it catches ghost kitchens more often, because the units are frequently in industrial zones where the answer is less obvious than it is on a retail strip.
The honest comparison
Against a restaurant of similar size, a delivery-only kitchen typically costs less per square foot — but not as much less as the missing dining room suggests, because the money moves into mechanical, refrigeration and storage.
Against a restaurant of similar revenue, the comparison is closer than people expect, because matching a restaurant’s output without its dining room means a denser, harder-working kitchen.
Where the model clearly wins is on rent and on the ability to run several concepts through one build. Where it disappoints is when someone budgets it as “a restaurant minus the dining room” and discovers that the remaining part is the expensive part.
Before you commit
- Confirm the zoning permits the specific use at the specific address.
- Size the exhaust and make-up air against the real equipment list, not the floor area.
- Plan storage — cold, dry and packaging — deliberately, and then add to it.
- Design the courier handoff as a real part of the plan.
- If it is multi-tenant, talk to the health unit before the drawings, not after.
The model works. It works best when it is budgeted as what it is — a production facility — rather than as a discount restaurant.
